def match(terms, labels, **kwargs):
"""Match given search terms (strings) to some given list of labels."""
verbosity = LOG.getverbosity(kwargs)
terms = ensurelist(terms, nonzero=True) # search terms
labels = ensurelist(labels, nonzero=True) # labels to match to
found = True
regex = kwargs.get('regex',
False) # use regexpr patterns (instead of glob)
incl = kwargs.get(
'incl', True
) # match at least one term; if incl=False ("exclusive"), match every term
start = kwargs.get('start', False) # match only beginning of string
LOG.verb(
"match: compare labels=%s -> searchterms=%s (incl=%s,regex=%s)" %
(labels, terms, incl, regex), verbosity, 3)
if not terms:
return False
for i, searchterm in enumerate(terms):
if not searchterm: continue
if not regex: # convert glob to regexp
#fnmatch.translate( '*.foo' )
#searchterm = re.sub(r"(?<!\\)\+",r"\+",searchterm) # replace + with \+
#searchterm = re.sub(r"([^\.])\*",r"\1.*",searchterm) # replace * with .*
searchterm = re.escape(searchterm).replace(r'\?', '.').replace(
r'\*', '.*?').replace(r'\^', '^')
if start and not searchterm.startswith('^'):
searchterm = '^' + searchterm
terms[i] = searchterm
if incl: # inclusive: match only one search term
for searchterm in terms:
for label in labels:
matches = re.findall(searchterm, label)
if matches:
LOG.verb(
" matched %r -> %r; return True" %
(label, searchterm), verbosity, 3)
return True # one search terms is matched
else:
LOG.verb(" not matched %r -> %r" % (label, searchterm),
verbosity, 3)
return False # no search term was matched
else: # exclusive: match all search terms
for searchterm in terms:
for label in labels:
matches = re.findall(searchterm, label)
if matches:
LOG.verb(" matched %r -> %r" % (label, searchterm),
verbosity, 3)
break
else:
LOG.verb(" not matched %r -> %r" % (label, searchterm),
verbosity, 3)
else:
return False # this search terms was not matched
return True # all search terms were matched
def shift(oldstr, shift, vars=["\w+"], **kwargs):
"""Shift all jet variable in a given string (e.g. to propagate JEC/JER).
E.g. shift('jpt_1>50 && met<50','jecUp',['jpt_[12]','met']) -> 'jpt_1_jecUp>50 && met_jecUp<50'
"""
verbosity = LOG.getverbosity(kwargs)
newstr = oldstr
vars = ensurelist(vars)
if re.search(r"(Up|Down)",oldstr):
print "shift: already shifts in %r"%(oldstr)
if kwargs.get('us',True) and len(shift)>0 and shift[0]!='_': # ensure underscore in front
shift = '_'+shift
for oldvar in vars: # shift each jet/MET variable
oldexp = r"\b("+oldvar+r")\b"
newexp = r"\1%s"%(shift)
newstr = re.sub(oldexp,newexp,newstr)
if verbosity>=1:
verbstr = ">>> shift: shift with %r shift: "%(newstr)
if len(newstr)<20:
verbstr += " %r -> %r"%(oldstr,newstr)
elif len(newstr)<35:
verbstr += "\n>>> %r -> %r"%(oldstr,newstr)
else:
verbstr += "\n>>> %r\n>>> -> %r"%(oldstr,newstr)
print verbstr
return newstr
def __init__(self, *args, **kwargs):
self.name = ""
self.title = ""
self.selection = ""
self.weight = ""
if len(args) == 1:
if isinstance(args[0], Selection):
self.name = args[0].name
self.title = args[0].title
self.selection = args[0].selection
else:
self.name = args[0]
self.title = args[0]
self.selection = args[0]
elif len(args) == 2:
self.name = args[0]
self.title = args[0]
self.selection = getselstr(args[1])
elif len(args) == 3:
self.name = args[0]
self.title = args[1]
self.selection = getselstr(args[2])
elif len(args) >= 4:
self.name = args[0]
self.title = args[1]
self.selection = getselstr(args[2])
self.weight = args[3]
self.title = kwargs.get('title', maketitle(self.title))
self.filename = makefilename(self.name)
self.filename = kwargs.get('fname', self.filename)
self.filename = kwargs.get('filename',
self.filename) # name for files, histograms
self.weight = kwargs.get('weight', self.weight)
#if self.selection=="":
# LOG.warning('Selection::Selection - No selection string given for %r!'%(self.name))
self.context = getcontext(
kwargs, self.selection) # context-dependent channel selections
self.only = kwargs.get('only', [])
self.veto = kwargs.get('veto', [])
self.only = ensurelist(self.only)
self.veto = ensurelist(self.veto)
def __init__(self, name, *args, **kwargs):
strings = [a for a in args if isinstance(a, str)]
self.name = name
self.name_ = name # backup for addoverflow
self.title = strings[0] if strings else self.name
self.filename = makefilename(self.name.replace('/', '_'))
self.title = kwargs.get('title', self.title) # for plot axes
self.filename = kwargs.get('fname', self.filename)
self.filename = kwargs.get('filename',
self.filename) # name for files, histograms
self.filename = self.filename.replace('$NAME', self.name).replace(
'$VAR', self.name) #.replace('$FILE',self.filename)
self.tag = kwargs.get('tag', "")
self.units = kwargs.get('units', True) # for plot axes
self.latex = kwargs.get('latex', True) # for plot axes
self.nbins = None
self.min = None
self.max = None
self.bins = None
self.cut = kwargs.get('cut', "") # extra cut when filling histograms
self.weight = kwargs.get(
'weight', "") # extra weight when filling histograms (MC only)
self.dataweight = kwargs.get(
'dataweight', "") # extra weight when filling histograms for data
self.setbins(*args)
self.dividebins = kwargs.get('dividebins', self.hasvariablebins(
)) # divide each histogram bins by it bin size (done in Plot.draw)
self.data = kwargs.get('data', True) # also draw data
self.flag = kwargs.get('flag', "") # flag, e.g. 'up', 'down', ...
self.binlabels = kwargs.get('labels', []) # bin labels for x axis
self.ymin = kwargs.get('ymin', None)
self.ymax = kwargs.get('ymax', None)
self.rmin = kwargs.get('rmin', None)
self.rmax = kwargs.get('rmax', None)
self.ratiorange = kwargs.get('rrange', None)
self.logx = kwargs.get('logx', False)
self.logy = kwargs.get('logy', False)
self.ymargin = kwargs.get(
'ymargin', None) # margin between hist maximum and plot's top
self.logyrange = kwargs.get(
'logyrange', None) # log(y) range from hist maximum to ymin
self.position = kwargs.get('pos', "") # legend position
self.position = kwargs.get('position',
self.position) # legend position
self.ncols = kwargs.get('ncol', None) # number of legend columns
self.ncols = kwargs.get('ncols',
self.ncols) # number of legend columns
#self.plot = kwargs.get('plots', True )
self.only = kwargs.get('only', []) # only plot for these patterns
self.veto = kwargs.get('veto', []) # do not plot for these patterns
self.blindcuts = kwargs.get('blind',
"") # string for blind cuts to blind data
self.addoverflow_ = kwargs.get('addoverflow',
False) # add overflow to last bin
if self.latex:
self.title = makelatex(self.title, units=self.units)
if 'ctitle' in kwargs:
for ckey, title in kwargs['ctitle'].iteritems():
kwargs['ctitle'][ckey] = makelatex(title)
if self.only:
self.only = ensurelist(self.only)
if self.veto:
self.veto = ensurelist(self.veto)
if self.binlabels and len(self.binlabels) < self.nbins:
LOG.warning("Variable.init: len(binlabels)=%d < %d=nbins" %
(len(self.binlabels), self.nbins))
if self.addoverflow_:
self.addoverflow()
if islist(self.blindcuts):
LOG.insist(
len(self.blindcuts) == 2,
"Variable.init: blind cuts must be a string, or a pair of floats! Got: %s"
% (self.blindcuts, ))
self.blindcuts = self.blind(*self.blindcuts)
self.ctxtitle = getcontext(kwargs,
self.title,
key='ctitle',
regex=True) # context-dependent title
self.ctxbins = getcontext(kwargs, args, key='cbins',
regex=True) # context-dependent binning
self.ctxposition = getcontext(kwargs,
self.position,
key='cposition',
regex=True) # context-dependent position
self.ctxblind = getcontext(
kwargs, self.blindcuts, key='cblind',
regex=True) # context-dependent blind limits
self.ctxymargin = getcontext(kwargs,
self.ymargin,
key='cymargin',
regex=True) # context-dependent ymargin
self.ctxcut = getcontext(kwargs, self.cut, key='ccut',
regex=True) # context-dependent cuts
self.ctxweight = getcontext(kwargs,
self.weight,
key='cweight',
regex=True) # context-dependent cuts